SHAMEFUL JOY

Posted on | by Ed

Full disclosure – not like you couldn't figure this out on your own – some of my less than laudatory attitude toward Nate Silver stems from professional jealousy. He is living quite literally a dream life and I am a professional failure who is going to die in rural central Illinois in one of those "We noticed a funny smell coming from the apartment" scenarios. He was smart enough, however, to realize that most people in the media and the public are ignorant about how math works and therefore they are likely to be extremely impressed by relatively basic statistical analysis. Yet despite the fact that what Silver does is not complicated to anyone who has some training in stats, it's often more complicated than it needs to be. It's an old academic trick – make something just complex enough that your audience won't be able to understand it and they will assume that you're correct because you're clearly smarter.

So it was that a man who called two of the easiest, least competitive presidential elections in the last few decades came to be seen as a modern oracle.

It is with no small amount of schadenfreude , then, that I watch the impending failure of his new, independent FiveThirtyEight website / media company. Proving once again that the New Media is about hits, hits, and trolling for more hits, Silver has elected to give an audience to a climate change "skeptic" to ensure plenty of outraged attention is directed toward his site devoted to "data-driven analysis." The shit has already hit the fan among some of his more high-profile liberal allies like Paul Krugman. What better way to emphasize one's slavish devotion to Data than to hire someone who has a decade long track record of consistently cherry-picking and manipulating climate data.

To paraphrase Bill Hicks, he might have been an artist at some point; if so, now he's just another set of holes at the capitalist gangbang. He'll find that making money without resorting to this kind of hackery will be considerably more difficult than, you know, calling a presidential election that one candidate ends up winning by 70 Electoral votes.

47 thoughts on “SHAMEFUL JOY”

  • Pat says:

    I forget whose theory on Nate Silver this was, but someone made a very astute comment that he seemed to be running away from everywhere he wasn't the smartest guy in the room anymore. He did what he did in the financial sector (as an analyst at UBS, if memory serves), then he went to Baseball Prospectus when baseball analytics were still closer to Joe Morgan than to Bill James. And PECOTA was genuinely a huge leap forward. But then the new generation of baseball geeks wound up being just as smart as he was, and maybe some were even smarter (Matt Swartz is one name to drop here), so he went to politics, where heaven knows the cretins still ruled. But now Sam Wang at Princeton is doing Nate Silver's schtick better than he ever did (I can't resist: to paint the lily, to Silver wrought silver). Maybe he just ran out of places to run….

  • J. Dryden says:

    I'll concede that Silver's character (among other things, his use of 'showmanship' while consistently claiming that what he was doing *wasn't* 'showmanship' was particularly annoying) isn't the greatest, and as Ed points out, what he does isn't difficult, but I don't think he can be blamed for the degree to which stupid people eager to puff up the mundane so that they don't have to do their jobs (the media) have elevated him above his deserving. Just saying. Crap decisions that he made, however, can be used against him.

    Still, I'll always–always–be grateful for the phenomenon he represents, which is the sect of people who, when confronted with, say, Bill Kristol's "gut," simply reply "Yeah, but, see: Math."

    Did he make "tough calls"? No. But he was right, and that means something, even when the bar for being right is insanely low.

    But mostly, I like Silver in large part because of the stupid people who hated him because his numbers didn't match their irrational desires.

  • Croooow! says:

    Pat-you're implying that Sam Wang is some Johnny come lately. He'd been doing similar work since before anyone heard of Nate silver.

  • Ed W. says:

    Whatever his faults might be, he's better than what came before him. Bill Kristol has had a 30+ year career as a pundit, with an accuracy rate of roughly .00000001%. If you want an "advanced" stat, his value over replacement pundit (VORP) is negative eleventy billion. Thomas Friedman apparently gets all of his ideas from interactions with Indian cab drivers, per Matt Taibbi. Peggy Noonan's sole function in life is to perform textual blow jobs on Ronnie Raygun's moldering corpse, for which she is paid real American dollars. So I guess what I'm saying is that he could be a lot worse.

  • Coises says:

    One line in Krugman’s column struck me:

    “What he seems to have concluded is that there are no experts anywhere, that a smart data analyst can and should ignore all that.”

    That’s a rather interesting (though not particularly novel) null hypothesis. Perhaps Mr. Silver will finally provide us with some conclusive evidence one way or the other.

    If he winds up demonstrating that a trained monkey with a spreadsheet can predict outcomes just as effectively as the most lauded experts in economics, sociology, criminology or climate science… I’ll be disappointed, but not very much surprised.

  • middle seaman says:

    Sam Wang, Nate Silver, Drew Linzer, and Simon Jackman built models trying to predict the 2012 elections. Nate Silver prior baseball involvement got him better press than others. The statistical models used aren't trivial and their precision isn't automatic.

    Silver may have stumbled but he and the others proved that the days of single pools and hysterical media should be behind us.

  • US in the UK says:

    I agree with Dryden, "I like Silver in large part because of the stupid people who hated him because his numbers didn't match their irrational desires." [See Bush's Brain's meltdown]

    I'm a quants geek and I enjoyed his book as a light-Quants thriller (not a growing genre) although I will confess it did make me a little jealous – "Gezus, I can do this.." But he got there first.

  • wetcasements says:

    "a man who called two of the easiest, least competitive presidential elections in the last few decades came to be seen as a modern oracle"

    But in the context of Peggy Noonan drunkenly consulting her inner fee-fee's or Joe Scarborough feeling the Mittmentum in his shrivelled balls, Silver was a qualitative step up from the usual talking heads. He caught a lot of shit for basically calling them out as useless.

    As for 538, I've checked it out twice. Very dull. There are plenty of interesting bloggers out there who would probably use a solid paycheck, and he didn't hire any of them.

    So I give him credit where credit is due re: 2012 (Markos had a more accurate predication, fwiw) but yeah, 538 is not looking so hot.

  • c u n d gulag says:

    Nate Silver lost me when he said that Ross Douthat was one of the few major Op-ed who is "someone who shows some originality."

    "Originality?!?!?!?!?!?!"

    Douche-hat's a Catholic version of Bobo Brooks.
    And just as inane, and just as vapid.

    And why not mention UpChuck Todd, as one of the few TV yakkers who is "someone who shows some originality?"

    ZOINKS!!!!!!!!!!!!

    After this, I think I'll stick to Sam Wang.

  • Gordon Guano says:

    My Impression of a Liberal Missing The Point:

    "Another set of holes at the capitalist gangbang"? Go back to RedState, homophobe!

  • Carter says:

    The idea that a consistent showing of 55% in a poll does not translate into a 55% chance of winning is important to convey to the public, and the fact that anyone with a basic education in statistics, or even a good general understanding of probability and math could point that out is beside the point, as they don't just let anyone on TV or write for the New York Times. He's pretty well hated because nothing scares people with backgrounds in the soft humanities like a falsifiable hypothesis, hence he did not "fit in with the culture" of the NYT.

    He's self-serving, but so is literally every public media figure involved in politics. He's raised the level of discourse, and so I'll read his blog. But yeah, new media is the death of quality, and I don't have high hopes for how it's going to go for him. I imagine that the only way he'll be able to stay in the black is to take money from campaigns, and then whatever credibility he has will be shot.

  • Rich says:

    If you read Silver's book, you'll know that he's a bit of a jut or at least someone who is pretty unworldly in many ways. Rather than regulate the financial system, he wants us to use the market to control debt management by turning our lives into an internet poker game (I'm over simplifying, but the obvious inspiration was his internet poker experience). Silver misses the point than normal people can't spend all their time playing the equivalent of internet poker and that the financial services industry already operates this way, which hasn't always gone well for it or the general public.

    Silver had one good idea and it builds on decades of of rather overlooked research on decision making. "Experts" ignore disconfirmatory evidence and often do less well than non-experts or chance alone. Actuarial data (in this case well design polls) are better predictors than say Charles Krauthamer, which is not exactly news. It isn't earth shattering but it's the sort of thing liberal arts majors like David Brooks are ill equipped to do.

    Given that Ezra Klein, a new generation's Tom Friedman, has hired a contrarian gay guy (a sort of road company Andrew Sullivan), I'm guessing that the overpaid schmucks who consult on start-ups are advising people (including Silver) to include some nutcase as a click driver in the tradition of Andrew Sullivan and Camile Paglia

  • lfv says:

    He made his bones using experts' data to make a predictions better than non experts. He looked great because no one actually listened to experts on elections, just pundits. Now he is doing something a bit different, not making predictions. And he seems to be consciously ignoring experts.

    I work in research. Statisticians are a great help, as support for the experts in the field. It is a joke to think they can function completely independently, let alone dabble in dozens of areas that all require deep expertise.

  • Leslee says:

    Although I agree with your overall assessment of Nate Silver, I could have done without the icky "gangbang" reference.

    Was that REALLY necessary?

  • CaptBackslap says:

    wetcasements: I'm in agreement overall, but can we get rid of 'fee-fees' forever? I've never, ever read that and thought less of the target, rather than the writer.

  • Barry says:

    Coises Says:

    "One line in Krugman’s column struck me:

    “What he seems to have concluded is that there are no experts anywhere, that a smart data analyst can and should ignore all that.”"

    IIRC, Krugman's idea could be summed up as: Nate Silver arbitraged between two fields – social science (polling, poli sci, etc.) and punditry. The former had a great deal of actual knowledge; the latter, none (aside from how to get paid for knowing nothing else). However, everybody knew what the pundits said, while what the real experts said was not well known. Silver was able to leverage this quite nicely.

    In fields like climate science, he's for some f-ing reason going in an looking at the real experts and the hack, liars, frauds and wh*res, and mixing them up.

  • Hazy Davy says:

    Mostly agrees with J. Dryden.

    I could (and have) develop predictive regressions and simulation models for football and stocks.

    I threw up my hands, trying to reconcile lots of polling data with varying types, magnitude, and directions of potential inaccuracy.

    Nate Silver built the model. That it came to an easy conclusion that was achievable with much simpler models isn't a reason to dismiss his. Did his complexity result in greater confidence in that result? I really do think so. And it was interesting for this non-statistics wizard. [Also, didn't he accurately call every single state? I mean, I miscalled a couple, I think…]

    I think he's made a mistake with the climate-hole guy, and it'll test his PR and self-promotion acumen.

    But even though I like you, Ed, I also admire Nate Silver, the good he's done for convincing people that data analysis is valid (and, for many of the mouth-breathing masses, gave them tools to evaluate other pundits), and what I think was a really interesting ex ante simulation of an easily-called election.

  • Anubis Bard says:

    I enjoyed Silver's shaming of the punditocracy during the election. But choosing a climate-change contrarian is an odd and annoying choice. I'm not sure if he's just going in for cheap controversy or whether he's ignorant about climate science. I suspect that Krugman has a point and Silver is intimidated by the actual numbers in the science, so he's chosen a mostly numerate gadfly who can take potshots at some of the summarizations. But unlike his use of polling data to clarify electoral matters, he's chosen someone who uses data purposely to bring less clarity to the topic at hand. It's too bad, because he's squandered all the credibility he had with me and many others.

  • Andrew says:

    What was amazing about Nate Silver was not that he predicted the overall outcome of the US presidential elections of 2008 and 2012 but that he predicted 50 of 51 jurisdictions correctly in 2008 and 51 of 51 correctly in 2012.

  • Jado says:

    The thing that always impressed me about nate Silver with regard to politics is that he was willing and able to resist the huge tide of resentment and influence and pressure that the Right imposed because they didn't like his analysis. I wasn't impressed that he was right – I was impressed that he never hemmed or hawwed while being right. He let the chips fall, and he let the right accuse him of "skewed polling", etc.

    Now he is cashing out with the webhits. It's always a shame when the champs take the mob money, but the name of the game is DOUGH-RE-MI, I guess.

    Thanks for the last few elections, Nate. Sam Wang can take it from here.

  • Anonymous says:

    @Andrew:

    That is not as impressive as it looks. Wang also went 101/102 in the same time span, and so long as you're following politics at all, your floor is high 80s or low 90s out of 102. Really, the true tossups were Indiana in 2008 and Florida in 2012, and both Wang and Silver went one for two. So long as you were looking at polling data, you would have gotten at least 100/102. Silver's strength is that he is a good marketer, not that he provides insight no one else can.

  • sluggo says:

    @ ED
    Love the gang bang reference. I can get sanitized language anywhere, speaking of which, Opiate of the Asses was a better tag line. I especially liked it when Rachel Maddow pixellated it.Remember the commitment to pants shitting.

  • BLOZAR says:

    The '…capitalist gangbang…' reference is a Bill Hicks (RIP) power metaphor from one of his comedic rants against the sleazy world of marketing (well worth looking up on youTube). Those of us familiar with Bill Hicks oeuvre enjoyed the reference not because we think sexual assault is OK, but because Bill Hick's made brilliant observations about how the values of integrity, honesty, fairness and decency in human relationships are so absent from venal, manipulative, dishonest, and exploitative relationships inherent to marketing.

    I imagined some of you prostrate on your fainting couches and fanning yourselves with embroidered hankies while squeaking out noises of protestation in response to an unexpected encounter with excessive vulgarity. I'm also imagining Ed's real sorry about that because he didn't set out to harm anyone (except maybe Nate Silver's feelings), but I could be wrong.

  • Drew says:

    @Blozar What are you talking about? I think you are confusing a gangbang with gang rape. Gangbangs, while, uh…you know, can be very much consensual.

    And if you don't like a blogger's language, read another blog. Nobody cares what you think about crude language.

  • Jude says:

    Yes, Drew. Thank you. Who doesn't enjoy a good consensual gangbang once in a while? They're what makes life worth living.

  • Nameless Poet says:

    Do you want to take your mother to a gangbang,
    Yes you do
    It's been a long time since she's had a screw
    But when she was young
    And in her prime
    She used to gangbang all the time.

    Thank you and good morning.

  • Alan C says:

    I'm just amazed how little being right (as was Silver in the elections) counts for in some quarters.

    When Silver's method for predicting the election was first publicized, he certainly didn't strike me as Statistics Jesus. "Oh, I see, he basically took all the polls and averaged them. That makes sense."

  • Drew says:

    Jude-it's not my fault you don't know what the word means. Some people might like getting banged by multiple people at once. Words have meanings, maybe stop imputing your own definition onto them.

  • Khaled says:

    I see everyone who is mortally offended by the word "gangbang" doesn't follow Gin and Tacos on Facebook. I hate to break it to you, but if you're going to be offended by a few swear words and some rough language, maybe a political website called "Gin and Tacos" isn't your cup of tea.

  • HoosierPoli says:

    Silver made his name calling the democratic PRIMARIES using data instead of conventional wisdom from Meet the Press, apparently something no one had ever done before.

    Also, economists like climate skepticism because inside of every economist is a raging contrarian who just likes to find ways to argue the inarguable.

  • pks says:

    I rather liked "another set of holes at the capitalist gang bang." A rather graphic and crude description of a fairly graphic and crude phenomena.

  • truth=freedom says:

    I came here to point out that hiring Pielke, while not particularly good from the perspective of those of us who think the collective pants-shitting ought to be reserved for the consequences of climate change that are coming *real* *soon*, isn't the unmitigated denial of climate reality that ThinkProgress portrays it as. Pielke's beef is (as best as I can tell) with the perception of urgency and the certitude that any climate changes will be disastrous. Now, I think he's really not paying attention, but he's not a "denier". He completely acknowledges that climate is, in part, driven by CO2, and that humans are doing their part to drive CO2 levels higher. He just doesn't think we should be spending any money to fix that. So, yeah, he's a total tool in his own way (I wonder if he has kids?), but he's not a "denier."

    After arriving here (and scanning the comments for previous mentions of this issue), I have to admit a certain discomfort with the awareness level of various commenters whose names I don't recognize. If the 'gang-bang' reference was "a bit much", perhaps G&T isn't for you? Even a cursory reading of the historical entries of this blog show a devotion to crude, violent, and even occasionally sadistic imagery as a way to get the point across (and to provide humor value). I for one welcome the unsanitized version of this blog.

  • Giselle Welschmeyer says:

    Write more, thats all I have to say. Literally, it seems as though you relied on the
    video to make your point. You definitely know what youre talking about, why throw
    away your intelligence on just posting videos to your blog when you could be giving us something enlightening to
    read?

Comments are closed.